Smith

I agree this is a pointless debate as you don't understand how correlation or confidence intervals work, so let's leave it there. Unfortunately this has derailed a far more interesting discussion we could all have had about the discrepancy itself.

The two lines are bound to reasonable correlated because they're closely related: one being an adjusted version of the other. But crucially you can see from the graph that the gap between the two lines is not constant. What Charlie and Alex have found is that over time the gap between raw and published data is getting larger. This means prices do not appear to be falling when you look at the adjusted figures, but they do if you look at the raw data. I'm not sure anyone is claiming this is a "cover up" exactly, more that whatever they're doing to adjust the raw data is leading to over reporting.

What makes you think there's a strong correlation between the index values and the raw data?

I'm saying that, based on bearishonhouses calculations, it's legitimate to base year-on-year calculations on the initial September 2023 release, because this isn't likely to change materially once all the figures are in. From your post above it doesn't sound like you've understood this, which is just going to cause confusion so we'll end up talking about what you don't understand, rather than the main issue. Just like we are now.

Hang on, please don't confuse this issue as it will make people who don't understand the simple stats here suspicious again. No one is suggesting that @bearishonhouses is comparing the index to raw data. That's not what they said and not what I said. Please go back and reread so you can edit your post.

Totally understand your position. I guess a lot of the difference in perspective depends on whether we have a generally positive or negative attitude about humanity and the way things are likely to go!

That's not the point I'm making. Just that it's hard to take someone seriously when they start talking about 15-minute cities and some vague 'plan' to price out the common man. Anyone who has half a brain and has taken the time to look at what 15-minute cities are actually about, rather than believing the conspiracy nonsense spouted about them, would understand that.

This sounds more like conspiracy theories than capitalism.

Thanks very much for taking the time to dig out these figures and do the calculations! If I'm reading it correctly then your analysis shows that the final published values (based on all transactions) for HPI do not historically vary by more than 2% from the initial published values (based on initial 10% or so of transactions). This means that criticisms of Charlie and Alex's methodology based on the fact that they're using the initial published values for September 2023 are invalid. As they say in the video, even if the final published values vary by 2% this barely alters the magnitude of the price drops they've uncovered. For example, the raw figures show that sale prices of flats and terraced houses have dropped on average by 19.6% in a year. Once the full data set is available, based on what you've found, the final figure is extremely unlikely to end up being less than a 17% drop or more than a 22% drop. This is the point Alex is making in the video when he talks about confidence limits. He's saying that whilst the limited number of data points for September 2023 makes it impossible to be totally accurate, the magnitude of what they've found is not likely to change.

Interesting. I'd like to see raw numbers compared between, say, June '22 and June '23, to see whether a proportion of those huge drops is evident. I may try to find these numbers over the weekend, if I get time. [APOLOGIES for not quoting anyone in my replies; this is still impossible to do on Android as ads cover the pop up.]

If the drops in non-adjusted figures are as dramatic as Charlie and Alex are claiming, then surely we could go back a few months and we should see the bulk of the percentage falls.

Still unusable out here in the sticks, sadly.

Likewise, my experience of being in the market as a buyer is that there's a LOT of reductions and very few buyers. Estate agents are falling over themselves to get buyers to make offers, even if they're low. This will of course vary by region, but where I am (cheap part of Hampshire) this is very obviously the case.

Confidence intervals are used to work out a likely range for the population mean (in this case, all September transactions) from the sample mean (in this case, the September transactions we know about so far) and standard deviation. So there's nothing wrong with this approach in principle. The issue is whether it's accurate to say this gives a range of only 2% in either direction. Rather than calculate this mathematically, it would be easier in this case to look at initial vs. final data from previous months and see how much the mean typically changes. If the final mean is never more than a few thousand different from the initial mean then their analysis is fair.

I also noticed that Alex's use of the term confidence intervals was a bit dubious. But the point he was making is nonetheless clear: that the initial 10% or so of data points posted by Land Registry for a given month have proved to be representative of the whole population, plus or minus a few percentage points. Do we know if this claim is accurate? If so, then I think they're analysis of the means and medians is fair.

Goes to show the importance of looking at the raw data, rather than being swayed by the commentary. In this case, the raw Land Registry figures show that sold prices have dropped by 12% in a year (19% for flats and terraced houses).

I don't live in London.

Am I supposed to want other people to suffer, then? Not sure I need that. I'll be happy with the end of the asset bubble and prices returning to sensible levels. Pretty sure that can happen without the need to turf pensioners out on the street. In fact, from where I'm standing it's already happening. If you want to feel better then maybe watch Moving Home with Charlie's video about 'The Discrepancy'. That might give you another perspective on what the data shows.

As it happens, I also sold and ended up renting (like you, not specifically as a financial decision). But unlike you I'm feeling pretty good about how things are turning out. 5+% interest on savings and property prices falling. Give it a year and we'll have more to spend on a better property. What's not to like!

I know you're just trying to have some fun with your fake profile, but all the same I'll reply as if you were serious: you'd do well to watch the video immediately above your post, and see what you think then about what the data shows.

You sound angry, not sure why. Auction properties can be good value if you've got cash, especially if you want something a little different or a bit of a project. No dicking around involved. No, I don't think all properties have fallen 20-30% already, which is why I didn't say that. But they've certainly fallen more than this one index suggests. And others on here have already explained to you why the index is not showing that reality. And by the way, it's not "30% off", it's 30% less than the peak. Not the same thing at all.

Not really true. It's currently possible to get over 5% easy access, which is about as high as savings rates have been in the last 15 years.

Interest only mortgages don't end; that's the problem.

What sounds like a rather desperate argument is attempting to cling on to one index, ignoring all criticism of its methodology, in the face of reality. I could buy the sort of house I want for 20-30% less than I could have bought the same house for in summer '22, allowing for reductions in asking price and the likelihood of a low offer being accepted. That's the truth of the matter.

House prices have been falling steadily now for over a year. This is obvious if you're out there looking and can see all the reductions in asking price, let alone allowing for a much lower offer being accepted in this market. If the data is telling you otherwise then get out and take a look for yourself.